Provider and facility factors in relation to receipt of dose-reduced chemotherapy among patients with breast cancer.

310 Background: Little is known about how provider and facility factors relate to patients’ receipt of dose-reduced adjuvant chemotherapy. Methods: In a cohort of 8,581 stage I-IIIA women with breast cancer treated with chemotherapy at Kaiser Permanente Northern California between 2006 and 2019, we evaluated the associations between provider- and facility-level factors related to chemotherapy dosing with a primary interest in the first cycle dose reduction (FCDR). Here, FCDR is defined as a first cycle dose proportion <90% of that expected based on the intended regimen, reflecting clinician intent at the outset of chemotherapy. Provider-level factors examined included provider gender, age, race, Hispanic ethnicity, years since completion of medical school, volume of breast cancer patients treated per year, and patient-provider racial concordance; we also examined practice size and rurality. Prevalence ratios (PR) were calculated using generalized linear models of the Poisson family with a log-link function and robust standard errors, accounting for significant patient factors and clustering at the provider level. Results were further mutually-adjusted for significant provider- and facility-level factors to ensure independence of associations. Results: Patients treated by providers practicing for longer (30+ years since medical school vs <10 years) were more likely to be dose-reduced at the outset of chemotherapy (PR: 1.94; 95% CI: 1.15-3.25; p-trend: 0.02). Patients treated by providers in a larger practice (10+ vs ≤5 oncologists) were less likely to be dose reduced (PR: 0.53; 95% CI: 0.34-0.84; p-trend: 0.03) as were patients treated by providers treating a higher volume of breast cancer patients (30+ vs ≤15 patients/year: PR 0.64 (95% CI: 0.41-0.99; p-trend: 0.03). Other factors examined were not significantly associated with use of FCDR. Conclusions: These results indicate provider years of experience, as well as practice size and the volume of breast cancer patients treated per year are significantly associated with first cycle dose reductions in chemotherapy. This is a complex, multi-factorial issue, and these results indicate that provider- and provider- and facility-level factors should be considered when evaluating sources of heterogeneity in chemotherapy delivery.